Design optimization methods using high-fidelity computational fluid dynamics simulations are becoming increasingly popular in the area of aerodynamic design, sustaining the desire to make these methods more computationally efficient. Such design strategies typically define the aerodynamic product using a parametric model of the geometry, but this can often require a large number of design variables, increasing the computational cost. This thesis proposes that a parametric model of aerodynamic flow features, rather than geometry, can be a parsimonious method of representing designs, giving a reduction in the number of design parameters required for optimization. The parameterization of flow features is coupled with inverse design, in order to recover the corresponding geometry. While an expensive analysis code is used in evaluating design performance, computational cost is reduced by using a low-fidelity code in the inverse design process. This newly presented method is demonstrated using four case studies in 2-D airfoil design, in which the parameterized flow feature is the surface pressure distribution, and two case studies for 3-D wing design, in which the spanwise loading distribution is parameterized. These strategies are consistently compared against a benchmark design search method which uses a conventional parameterization of the geometry. The two methods are described in detail, and their relative performance is analysed and discussed. The newly presented method is found to converge towards the optimum design significantly more quickly than the benchmark method, providing designs with greater performance for a given computational expense. A parameterization of flow features can generate designs with higher quality and detail than a geometry-based method of the same dimensionality.

No description available.

No description available.

No description available.

Increasingly, large volumes of aircraft data are being recorded in an effort to adapt aircraft maintenance procedures from being time-based towards condition-based techniques. This study uses techniques of artificial intelligence and develops mathematical models to analyse this data to enable improvements to be made in aircraft management, affordability, availability, airworthiness and performance. In addition, it highlights the need to assess the integrity of data before further analysis and presents the benefits of fusing all relevant data sources together. The research effort consists of three separate investigations that were undertaken and brought together in order to provide a unified set of methods aimed at providing a safe, reliable, effective and efficient overall procedure. The three investigations are: 1. The management of helicopter Health Usage Monitoring System (HUMS) Condition Indicators (CIs) and their analysis, using a number of techniques, including adaptive thresholds and clustering. These techniques were applied to millions of CI values from Chinook HUMS data. 2. The identification of fixed-wing turbojet engine performance degradation, using anomaly detection techniques, applied to thousands of in-service engine runs from Tornado aircraft. 3. The creation of models to identify unusual aircraft behaviour, such as uncommanded flight control movements. Two Chinook helicopter systems were modelled and the models were applied to over seven hundred in-service flights. In each case, the existing techniques were directed toward a condition-based maintenance approach, giving improved detection and earlier warning of faults.

In this work physics-based modelling of bluff-body noise was performed with application to landing gear noise production. The landing gear is a primary contributor to airframe noise during approach. Noise is primarily generated from the unsteady pressures resulting from the turbulent flow around various components. The research was initiated in response to the need for an improved understanding of landing gear noise prediction tools. A computational approach was adopted so that the noise generating physics of the problem could be captured. Governing laws were solved numerically to predict the noise source characteristics and the resulting acoustic far-field. Three-dimensional compressible Navier-Stokes simulations were performed to solve the unsteady turbulent near-field flow and the acoustic analogy was used to predict the resulting far-field acoustic pressure. The flow solver included a high-order computational aeroacoustics code adopting large-eddy simulation, whilst a Ffowcs Williams and Hawkings solver was used for the acoustic prediction. Circular cylinders in various configurations were selected to represent basic landing gear struts and results were used to form a modelling database. Initially, cylinders at various Reynolds numbers were investigated in cross-flow to determine the noise characteristics of a simple model strut. The work was extended to investigate the effect of strut alignment to the flow by simulating cylinders in yaw. The effect of yaw was shown to modify the peak level and frequency of far-field noise spectra. Component interaction effects were then investigated by simulating cylinders in tandem arrangements. The resulting aerodynamic and far-field noise characteristics were shown to be complex and extremely sensitive to the separation distance between the cylinders. Finally, a prediction model was developed and validated by comparing predictions against theory and measurements of the noise radiated by a simple two-wheel landing gear model. The results demonstrated the capability of the model to accurately predict correct spectral and directivity characteristics.

According to one estimate, about 40 percent of the driving population will be over the age of 60 by the year 2020 in the UK and currently, several hundred thousand drivers with dementia hold driving licenses. The number of motor vehicle crashes per unit distance of automobile travel is “U”-shaped, with risk increasing slightly between the ages of 55 and 60, but risk increasing with each successive five-year interval. Some individuals who have mild dementia possess sufficient driving skills to be designated as fit drivers. The most challenging assessment and decision for the physician/licensing authority as regards fitness to drive lies in drivers who are questionably demented or are in a state of very mild dementia. In the absence of a reliable standard protocol, some clinicians make judgment based on selfreporting, which has risks associated with it as lack of insight and judgment are potential common traits of the population experiencing cognitive decline. Seldom is recourse made by health professionals to on-road assessment as a first alternative as it requires a fee and such testing centers are not readily available everywhere. This research addresses this issue of the identification of cognitive tests that can be used to assess an individual’s ability to drive and especially of those individuals that are questionably demented and are the most difficult to identify. A younger and an older group consisting of 56 drivers in total were administered nine different cognitive tests and two drives (Drive-I and Drive-II) on the STISIM driving simulator. The cognitive test ufov3 (involving the identification of a central target and simultaneously the radial localization of a peripheral target embedded in distracter triangles), which is the third subtest of the UFOV (Useful Field of View) test showed the highest discriminating ability in separating “poor-drivers” from “not-poor-drivers”, with 92.86 % of the drivers correctly classified. The next best discriminating ability in decreasing order of strength was that of dichotic listening test, trail making test, rey-copy test and paper folding test. Also, age was found to be an excellent discriminator of “poor-drivers” and “not-poor-drivers” with 91.07 % of the drivers correctly classified. A composite cognitive measure consisting of the sum of all nine cognitive tests was not a better predictor than the ufov3 test alone; overall it was still an excellent discriminator, classifying 89.29 % of drivers correctly. The commonly recommended Clock Drawing test and the Trail Making test did not emerge as significant predictors of driving ability. A general driving skills linear model for prediction purposes was derived that explained 59 % of the variation in a general driving performance index with the ufov3 test, the dichotic listening test and the rey-recall test as significant predictors. Recommendations are made as to how this test should be used to screen potentially at risk drivers.

No description available.

Direct numerical simulation was used to study various aspects of vortex ring evolution and interaction with a free surface. An investigation of a single unbounded vortex ring at various Reynolds numbers and core thicknesses showed qualitative dierences between the evolution of thin- and thick-core rings, leading to a correction factor to the classical equation for the ring translational velocity. The obtained linear modal growth rates were compared with previous work, highlighting the role of the wake in triply periodic numerical simulations. The transition from a laminar to a turbulent ring is marked by the rearrangement of the outer core vorticity into a clearly dened secondary structure. The onset of the fully turbulent state is associated with shedding of the structure in a series of hairpin vortices. A Lagrangian particle analysis was performed to determine the ring entrainment and detrainment properties and to investigate the possibility of an axial flow being generated around the circumference of the core region prior to the onset of turbulence. The orthogonal interaction of laminar, transitional and turbulent rings with an initially undesturbed free surface was investigated. At small depths, the expanding ring is unstable to the Crow instability but its dominant mode is predetermined by the prior development of the Widnall instability. The presence of opposite-signed vorticity, due to surface curvature, aects the ring dynamics at the surface. The interaction of a transitional ring modies the surface displacements, refecting the structure of the ring below. The secondary structure associated with a transitional ring reconnects to the surface in addition to the inner core. In the presence of the surface, the turbulent ring nds greater coherency of the core due to stretching and aligning of vorticity laments. The addition of a planar surface wave field modied the ring interaction greatly due to the higher surface curvature and associated surface vorticity. The ring expands asymmetrically and even rebounds locally if sufficient opposite-signed vorticity is generated. The ring diffracts the surface wave field and the generation of secondary small-amplitude waves was noted.

Two-dimensional (2-D) cavity flow physics and oscillation control were investigated through solutions of the Reynolds-Averaged Navier-Stokes equations coupled with a two-equation k-ω turbulence model. Effects of the leading edge modifications including compression ramps, expansion surfaces and mass injection on supersonic cavity flow oscillation were investigated. Different flow mechanisms were observed at Mach 1.5 and 2.5. The study proposed a different explanation of the flow control mechanism when mass injection was used. An optimal mass injection rate was identified. A further improvement on the 2-D model was made by considering the effect of the turbulent viscous sub-layer using the Integration-to-the-wall boundary condition. The results confirmed that the shear layer instability reduced gradually from Mach 1.5 to 3.5. An improvement in SPL prediction was achieved. Dominant modes were also correctly predicted. The capability of the model was extended for the investigation of the 3-D compressible unsteady turbulent flow physics. It was validated against a hypersonic symmetric corner flow. The turbulent effect was modelled by a two-equation k-ω turbulence model. A laminar cavity flow oscillation at Mach 1.5 was predicted. The result showed a self-sustained pressure oscillation. The predicted pressure oscillation was dominated by the second mode and its frequency was 5702H z which was close to the measured value of 5900H z. The SPL discrepancies with the measurements were within 2.3dB. A secondary symmetric flow pattern inside the cavity displayed a 3-D effect and showed the effect of the side wall within a spanwise distance of 2D. Further simulation of the turbulent cavity flow at Mach 1.5 showed a weak pressure oscillation, which indicated the RANS is sensitive to the choice of turbulence model. However, the surface flow pattern and surface pressure distribution were consistent with the measurements. The strongest surface pressure oscillation was observed near the rear corner on the centre plane. A secondary symmetric flow pattern also existed. A spanwise wave was found in the cavity. At the trailing plate, a flow separation was formed in the laminar boundary layer but was not observed in the turbulent boundary layer.